Intelligent Prediction Method and Device for Turbomachinery Support Stiffness

Abstract

The present invention provides an intelligent prediction method and device for the support stiffness of a turbomachine, belonging to the technical field of mechanical structures. The method includes: establishing a reduced finite element model based on a real turbomachine foundation support model; using the Latin hypercube sampling method to construct a data sample set of the support stiffness and damping ratio of the reduced finite element model; performing finite element analysis on each data sample in the data sample set to generate dynamic stiffness curves of the support structure in the vertical and horizontal directions; constructing an intelligent prediction model for the turbomachine support stiffness according to the data sample set and the dynamic stiffness curves of the support structure in the vertical and horizontal directions; and performing intelligent prediction of the turbomachine support stiffness by using the intelligent prediction model for the turbomachine support stiffness. The present invention can accurately capture the characteristics of a complex dynamic system, provide more reliable technical support and decision-making basis, and further optimize equipment management and maintenance, and improve the overall performance and service life of the turbomachine.

Description

Background Art

[0002] As a key power equipment in industrial production, the operating stability and safety of turbomachinery are directly related to the efficiency and reliability of the entire system. The support stiffness of turbomachinery is an important parameter to measure the structural stability of the equipment, and its influencing factors are complex, including various physical properties such as support stiffness and damping ratio. In practical applications, the operating process of turbomachinery may face complex and changeable working conditions, and the reasonable prediction of support stiffness is crucial for the evaluation and maintenance of the equipment operating state.

[0003] In the prior art, traditional finite element analysis methods are mostly used to analyze and predict the support stiffness of turbomachinery. These methods usually require a large amount of experimental data and complex calculation processes to obtain key parameters and dynamic stiffness curves. However, with the improvement of the complexity of industrial equipment, the traditional methods have significant defects in the following aspects: First, the accuracy of the finite element model depends on the input of high-precision parameters. However, under actual operating conditions, the diversity and uncertainty of the input data are often difficult to fully cover, resulting in insufficient reliability of the prediction results; Second, for the dynamic response analysis of multiple variables, the calculation efficiency of traditional methods is low, and it is difficult to meet the real-time requirements, especially when dealing with complex changes in support stiffness and damping ratio; Third, when the existing technology calculates the support stiffness in different directions (such as vertical and horizontal directions), it often cannot take into account both the global and local characteristics, resulting in limited accuracy of the analysis results.

[0004] In addition, for the prediction of the support stiffness of turbomachinery, the prior art has not yet achieved the best balance in model simplification and data sampling, and usually has problems of excessive consumption of computing resources and insufficient data coverage. Limited sample data cannot fully express the complex changes of support stiffness and damping ratio, thereby limiting the generalization ability of the prediction model. At the same time, in the absence of in-depth exploration of non-linear characteristics, it is difficult to further improve the prediction accuracy, especially in practical applications under multiple working conditions, prediction deviations are likely to occur, affecting the long-term stable operation of the equipment. Summary of the Invention

[0005] To overcome the problems existing in the related art, the present invention provides an intelligent prediction method and device for the support stiffness of turbomachinery.

[0006] According to the first aspect of the embodiments of the present invention, an intelligent prediction method for the support stiffness of turbomachinery is provided, and the method includes:

[0007] Based on a real turbomachinery foundation support model, a reduced finite element model is established;

[0008] Using the Latin hypercube sampling method, a data sample set of the support stiffness and damping ratio of the reduced finite element model is constructed;

[0009] Perform finite element analysis on each data sample in the data sample set to generate the dynamic stiffness curves of the support structure in the vertical and horizontal directions;

[0010] Construct an intelligent prediction model for the support stiffness of a turbomachine based on the data sample set and the dynamic stiffness curves of the support structure in the vertical and horizontal directions;

[0011] Use the intelligent prediction model for the support stiffness of a turbomachine to perform intelligent prediction of the support stiffness of a turbomachine.

[0012] In some exemplary embodiments of the present invention, based on the foregoing solution, using the intelligent prediction model for the support stiffness of a turbomachine to perform intelligent prediction of the support stiffness of a turbomachine includes:

[0013] Perform multi-scale feature extraction and transformation on the newly input support stiffness and damping ratio data to obtain a transformation result;

[0014] Perform a transformation operation on the transformation result to obtain a multi-scale feature vector;

[0015] Perform feature fusion on the multi-scale feature vector to obtain an intelligent prediction result for the support stiffness of a turbomachine;

[0016] Wherein, the intelligent prediction result for the support stiffness of a turbomachine includes the dynamic stiffness curves of the support structure in the vertical and horizontal directions.

[0017] In some exemplary embodiments of the present invention, based on the foregoing solution, performing multi-scale feature extraction and transformation on the newly input support stiffness and damping ratio data to obtain a transformation result includes:

[0018] Extract the different-scale features of the newly input support stiffness and damping ratio data;

[0019] Utilize the non-linear changes of multiple activation functions to transform the feature extraction result of each scale of the newly input support stiffness and damping ratio data into a new representation;

[0020] Perform weighted summation and non-linear transformation on the new representation to obtain a transformation result.

[0021] In some exemplary embodiments of the present invention, based on the foregoing solution, utilizing the non-linear changes of multiple activation functions to transform the feature extraction result of each scale of the newly input support stiffness and damping ratio data into a new representation includes:

[0022] ;

[0023] Wherein, represents the newly formed representation, is an activation function and , Represents the weights of the neurons in the th n layer corresponding to the l th neuron for connecting the newly input support stiffness and damping ratio data, is the corresponding bias, represents the value of the th B-spline basis function at , N is the number of B-spline basis functions, are the control points of the spline basis functions, represents the grid size at scale , represents the spline order of the B-spline curve at scale .

[0024] In some exemplary embodiments of the present invention, based on the foregoing solution, a transformation operation is performed on the conversion result to obtain a multi-scale feature vector including:

[0025] ;

[0026] wherein, represents the conversion result obtained at scale and , is an activation function and , represents the weights of the second neuron in the th layer for connecting the newly input support stiffness and damping ratio data n , is the corresponding bias, represents the value of the th B-spline basis function at , N is the number of B-spline basis functions, are the control points of the spline basis functions, represents the grid size at scale and represents the spline order of the B-spline curve at scale .

[0027] In some exemplary embodiments of the present invention, based on the foregoing solution, feature fusion is performed on the multi-scale feature vector to obtain an intelligent prediction result of the turbine machinery support stiffness, including:

[0028] ;

[0029] wherein, represents the intelligent prediction result of the turbine machinery support stiffness, represents the weights of the fusion layer, represents a multi-scale feature vector, represents the bias of the fusion layer.

[0030] In some exemplary embodiments of the present invention, based on the foregoing solution, the loss function of the turbomachinery support stiffness intelligent prediction model includes:

[0031] ;

[0032] wherein, represents the number of samples, represents a loss function for measuring the difference between the predicted value and the true value, represents the network model, represents the scale s at the m input of the th sample of the

[0033] According to a second aspect of the embodiments of the present invention, there is provided an apparatus based on the above turbomachinery support stiffness intelligent prediction method, including:

[0034] A model establishment module, configured to establish a reduced finite element model based on a true turbomachinery foundation support model;

[0035] A sample set construction module, configured to construct a data sample set of the support stiffness and damping ratio of the reduced finite element model by using the Latin hypercube sampling method;

[0036] A curve generation module, configured to perform finite element analysis on each data sample in the data sample set to generate dynamic stiffness curves of the support structure in the vertical and horizontal directions;

[0037] A prediction model generation module, configured to construct a turbomachinery support stiffness intelligent prediction model according to the data sample set and the dynamic stiffness curves of the support structure in the vertical and horizontal directions;

[0038] An implementation prediction module, configured to perform turbomachinery support stiffness intelligent prediction by using the turbomachinery support stiffness intelligent prediction model.

[0039] According to a third aspect of the embodiments of the present invention, there is provided an electronic device, including: a processor; and a memory, where a computer-readable instruction is stored on the memory, and when the computer-readable instruction is executed by the processor, the turbomachinery support stiffness intelligent prediction method in the first aspect is implemented.

[0040] According to a fourth aspect of an embodiment of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the intelligent prediction method for the support stiffness of a turbomachine in the first aspect is implemented.

[0041] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0042] In the embodiments of the present invention, by combining finite element analysis, data sampling technology, and an intelligent prediction model, the characteristics of a complex dynamic system can be accurately captured, providing more reliable technical support and decision-making basis, thereby optimizing equipment management and maintenance, and improving the overall performance and service life of the turbomachine.

[0043] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings herein are incorporated into the specification and form a part of the present invention, showing embodiments consistent with the present invention, and are used together with the specification to explain the principles of the present invention.

[0045] Figure 1 A schematic diagram of a system architecture showing an exemplary application environment of an intelligent prediction method and device for the support stiffness of a turbomachine to which the embodiments of the present invention can be applied;

[0046] Figure 2 A schematic flow diagram showing the intelligent prediction method for the support stiffness of a turbomachine according to some embodiments of the present invention;

[0047] Figure 3 (a) Schematically shows a dynamic stiffness curve graph in the vertical direction according to some embodiments of the present invention; Figure 3 (b) Schematically shows a dynamic stiffness curve graph in the horizontal direction according to some embodiments of the present invention;

[0048] Figure 4 A schematic structural diagram showing a multi-scale Kolmogorov - Arnold network according to some embodiments of the present invention;

[0049] Figure 5 (a) Schematically shows a schematic diagram of the loss change of each model under vertical direction training data according to some embodiments of the present invention; Figure 5 (b) Schematically shows a schematic diagram of the loss change of each model under horizontal direction training data according to some embodiments of the present invention;

[0050] Figure 6(a) Schematically shows the R coefficient index diagrams of CNN, ResNet, and KAN at multiple scales and MKAN in the dynamic stiffness prediction of the vertical direction test set according to some embodiments of the present invention; 2 Coefficient index diagram; Figure 6 (b) Schematically shows the MAE index diagrams of CNN, ResNet, and KAN at multiple scales and MKAN in the dynamic stiffness prediction of the vertical direction test set according to some embodiments of the present invention; Figure 6 (c) Schematically shows the RMSE index diagrams of CNN, ResNet, and KAN at multiple scales and MKAN in the dynamic stiffness prediction of the vertical direction test set according to some embodiments of the present invention;

[0051] Figure 7 (a) Schematically shows the prediction result diagram of the convolutional neural network CNN in Experiment 1 according to some embodiments of the present invention; Figure 7 (b) Schematically shows the prediction result diagram of the residual network ResNet in Experiment 1 according to some embodiments of the present invention; Figure 7 (c) to Figure 7 (e) Schematically shows the prediction result diagrams of the novel neural network architecture KAN at different grid sizes in Experiment 1 according to some embodiments of the present invention; Figure 7 (f) Schematically shows the prediction result diagram of the multi-scale Kolmogorov - Arnold network MKAN in Experiment 1 according to some embodiments of the present invention; True in the figure represents the true value;

[0052] Figure 8 (a) Schematically shows the R coefficient of CNN, ResNet, and KAN at multiple scales and MKAN in the dynamic stiffness prediction of the horizontal direction test set according to some embodiments of the present invention; 2 Coefficient; Figure 8 (b) Schematically shows the MAE of CNN, ResNet, and KAN at multiple scales and MKAN in the dynamic stiffness prediction of the horizontal direction test set according to some embodiments of the present invention; Figure 8 (c) Schematically shows the RMSE of CNN, ResNet, and KAN at multiple scales and MKAN in the dynamic stiffness prediction of the horizontal direction test set according to some embodiments of the present invention;

[0053] Figure 9 (a) Schematically shows the prediction result diagram of the convolutional neural network CNN in Experiment 2 according to some embodiments of the present invention; Figure 9 (b) Schematically shows the prediction result diagram of the residual network ResNet in Experiment 2 according to some embodiments of the present invention; Figure 9 (c) to Figure 9(e) Schematically shows the schematic diagram of the prediction results of the novel neural network architecture KAN in Experiment 2 according to some embodiments of the present invention under different grid sizes; Figure 9 (f) Schematically shows the schematic diagram of the prediction results of the multi-scale Kolmogorov - Arnold network MKAN in Experiment 2 according to some embodiments of the present invention; True in the figure represents the true value;

[0054] Figure 10 Schematically shows the schematic diagram of the intelligent prediction device for the support stiffness of a turbomachine according to some embodiments of the present invention;

[0055] Figure 11 Schematically shows the schematic diagram of the structure of a computer system of an electronic device according to some embodiments of the present invention;

[0056] Figure 12 Schematically shows the schematic diagram of a computer-readable storage medium according to some embodiments of the present invention. Specific embodiments

[0057] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0058] The terms used in the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The singular forms "a", "the" and "said" used in the present invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0059] It should be understood that although the terms first, second, third, etc. may be used in the present invention to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present invention, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "when" or "while" or "in response to determining".

[0060] Figure 1 Shows a schematic diagram of the system architecture of an exemplary application environment to which the intelligent prediction method and device for the support stiffness of a turbomachine according to the embodiments of the present invention can be applied.

[0061] As Figure 1 shown, the system architecture 100 may include one or more of terminal devices such as a desktop computer 101, a portable computer 102, a smart phone 103, a network 104, and a server 105. The network 104 is used to provide a medium for a communication link between the terminal device and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc. The terminal device may be various electronic devices with data processing functions, and the electronic device has a display screen, and the display screen is used to display the intelligent prediction result of the turbine machinery support stiffness to the user, including but not limited to the above-mentioned desktop computer, portable computer, smart phone, etc. It should be understood that Figure 1 the numbers of terminal devices, networks, and servers in

[0062] are merely illustrative. According to the implementation requirements, there may be any number of terminal devices, networks, and servers. For example, the server 105 may be a sub-server cluster composed of multiple sub-servers, etc.

[0063] In addition, it should be understood that the intelligent prediction method for the turbine machinery support stiffness according to the embodiments of the present invention may be configured as a software module. In some implementation scenarios, the intelligent prediction solution for the turbine machinery support stiffness of the present invention may be deployed separately to generate different prediction contents. In other implementation scenarios, the intelligent prediction solution for the turbine machinery support stiffness of the present invention may be deployed in other software as a functional module of the software, such as being deployed in the analysis software for the turbine machinery support stiffness. The present invention does not make special restrictions on the application manner of the intelligent prediction method for the turbine machinery support stiffness.

[0064] Next, the embodiments of the present invention will be described in detail.

[0065] As Figure 2 shown, Figure 2 is a flowchart of an intelligent prediction method for the turbine machinery support stiffness according to an exemplary embodiment of the present invention, including the following steps:

[0066] S210: Based on the actual turbine machinery foundation support model, establish a reduced finite element model;

[0067] S220: Use the Latin hypercube sampling method to construct a data sample set of the support stiffness and damping ratio of the reduced finite element model;

[0068] S230: Perform finite element analysis on each data sample in the data sample set to generate the dynamic stiffness curves of the support structure in the vertical and horizontal directions;

[0069] S240: Construct an intelligent prediction model for the support stiffness of the turbomachinery based on the data sample set and the dynamic stiffness curves of the support structure in the vertical and horizontal directions;

[0070] S250: Use the intelligent prediction model for the support stiffness of the turbomachinery to perform intelligent prediction of the support stiffness of the turbomachinery.

[0071] In S210, a reduced finite element model is established based on the actual turbomachinery foundation support model.

[0072] The turbomachinery foundation support model refers to a mathematical or physical model of the foundation structure used to support the turbomachinery. It usually describes how the foundation bears the loads generated by the turbomachinery (such as vibration, pressure, temperature changes, etc.) during operation. In the embodiments of the present invention, the turbomachinery foundation support model focuses on the support effect and support stiffness of the foundation on the turbomachinery. The support stiffness refers to the resistance ability of the foundation to displacement when subjected to a load. The support performance of the foundation is very important for the stability and operation effect of the turbomachinery.

[0073] The turbomachinery foundation support model can be established by measuring the deformation and stress of the foundation and combining the principles of structural mechanics.

[0074] The reduced finite element model refers to reducing the degrees of freedom (DOF) of the original finite element model through certain techniques, thereby reducing the computational amount. The reduced finite element model can be established by using methods such as modal reduction, principal component analysis reduction, or assumed element method reduction.

[0075] In the embodiments of the present invention, establishing a reduced finite element model based on the actual turbomachinery foundation support model further includes:

[0076] Add master nodes at six constraint positions on the foundation base of the actual turbomachinery foundation support model;

[0077] Establish a reduced finite element model according to the turbomachinery foundation support model after adding master nodes.

[0078] Here, the six constraint positions refer to the support positions of the turbomachinery foundation base, which correspond to the fixed points of the foundation. The foundation is connected to the surrounding environment (such as the ground, foundation) through these constraint points. These constraints are usually rigid supports (such as fixed supports, hinged supports, etc.), which limit the degrees of freedom of the foundation in specific directions. Adding master nodes at the six constraint positions of the foundation base means setting nodes at these constraint positions, and these nodes are responsible for bearing the constraint conditions. For a three-dimensional structure, there are usually six degrees of freedom (three translational degrees of freedom (displacements in the X, Y, and Z directions), and three rotational degrees of freedom (rotations about the X, Y, and Z axes)) to describe the motion of the nodes.

[0079] By adding master nodes, it is possible to control the displacement and rotation behavior of the turbomachinery foundation and establish its mechanical model.

[0080] On this basis, a reduced finite element model is established. When establishing the reduced finite element model, the geometric model of the foundation base can be created first, master nodes are set at the six constraint positions of the foundation, and then the element type (which can be beam elements, shell elements, solid elements, etc.) is selected, and the material properties are defined, etc.

[0081] By establishing a realistic support model of the turbomachinery foundation, the key features of the turbomachinery foundation support can be retained. While ensuring the calculation accuracy, it can reduce the resources and time required for calculation, providing an efficient data basis for subsequent predictive analysis; on this basis, using the reduced finite element method for simplified calculation can effectively reduce the calculation complexity and improve the calculation efficiency.

[0082] In S220, using the Latin hypercube sampling method, a data sample set of the support stiffness and damping ratio of the reduced finite element model is constructed.

[0083] Latin Hypercube Sampling (LHS) is an efficient statistical sampling method used to uniformly extract samples in a multi-dimensional space. Compared with the traditional random sampling method, LHS can ensure a uniform distribution within the value range of each variable and avoid the phenomenon that the samples are concentrated in a certain area. The LHS method can fully explore the parameter space of different variables in the reduced finite element model.

[0084] Support stiffness is a physical quantity that describes the resistance of a support system to external forces or deformations. It is usually expressed as the force required for unit displacement. Support stiffness can be defined in different directions (such as horizontal and vertical directions), and the common units are N / m or N·m / rad (for rotational stiffness). The damping ratio is a dimensionless parameter that describes the vibration attenuation ability of a system. It is usually defined as the ratio of the actual damping of a structure to the critical damping. It is closely related to characteristics such as the vibration attenuation and energy consumption of the system and is usually expressed as a percentage. The damping ratio ranges from 0% (undamped) to 5% - 10% (high damping).

[0085] The process of constructing a data sample set of the support stiffness and damping ratio of the reduced finite element model using the Latin hypercube sampling method can be as follows:

[0086] Based on actual design requirements, material properties, or existing experimental data, define the value ranges of the support stiffness and damping ratio. For example, the support stiffness can be taken from 1000 N / m to 5000 N / m; the range of the damping ratio is from 0.01 (1%) to 0.05 (5%).

[0087] Divide the parameter space. Divide the value range of each parameter into several equal intervals, and then randomly select a value from each interval. Suppose N samples are to be generated, then the interval of each parameter will be divided into N sub - intervals, and a sample point will be randomly selected from each sub - interval.

[0088] Generate samples. Use the Latin hypercube sampling method to uniformly select a sample point from each interval. For example, for N samples, N samples are drawn in the range from 1000 N / m to 5000 N / m, and N samples are drawn in the range from 0.01 (1%) to 0.05 (5%). Ensure independent random sampling during the drawing process.

[0089] Combine the N results drawn from the range of the support stiffness and the N results drawn from the range of the damping ratio to form the data sample set. In the present invention, N = 100.

[0090] In S230, perform finite element analysis on each data sample in the data sample set to generate the dynamic stiffness curves of the support structure in the vertical and horizontal directions.

[0091] Dynamic Stiffness describes the stiffness characteristics of a structure under dynamic loads (such as vibration and shock). Dynamic stiffness is frequency - dependent, that is, at different frequencies, the stiffness value of the support structure will be different. The dynamic stiffness curve is a graph with frequency as the abscissa and dynamic stiffness as the ordinate.

[0092] In a finite element analysis software, the support stiffness and damping ratio can be applied as constraint conditions on the base nodes or contact surfaces.

[0093] On this basis, in order to obtain the dynamic stiffness curve, the frequency scanning method can be used to apply dynamic loads (such as sinusoidal excitation, impact load or random vibration) at different frequencies of the support structure. Then, according to the applied dynamic load and support conditions, the dynamic response of the support structure at different frequencies is solved. The dynamic stiffness is calculated based on the response results. Then, a curve of the dynamic stiffness varying with frequency is plotted, as shown in Figure 3 Figure (a) and Figure 3 Figure (b).

[0094] Here, the calculation method of the dynamic stiffness can be:

[0095] ;

[0096] where is the dynamic stiffness, is the force applied at frequency and is the displacement response at this frequency.

[0097] Performing finite element analysis on each data sample in the data sample set to generate the dynamic stiffness curves of the support structure in the vertical and horizontal directions is actually to simulate the dynamic response of the support structure through a finite element model and calculate the dynamic stiffness of each sample at different frequencies. This process includes defining the model, applying loads, performing frequency response analysis, calculating the dynamic stiffness and generating the dynamic stiffness curve.

[0098] In S240, according to the data sample set and the dynamic stiffness curves of the support structure in the vertical and horizontal directions, an intelligent prediction model for the support stiffness of the turbomachinery is constructed.

[0099] Here, the data sample set can be used as the input data of some machine learning models, and the dynamic stiffness curves of the support structure in the vertical and horizontal directions can be used as the output data to train the machine learning models. The trained machine learning model is the intelligent prediction model for the support stiffness of the turbomachinery.

[0100] The present invention does not limit the specific structure of the machine learning model. For example, in some embodiments, the machine learning model can be a deep neural network, a convolutional neural network, a long short-term memory network, etc.

[0101] In the embodiments of the present invention, the machine learning model is a multi-scale Kolmogorov - Arnold network (MKAN) model, and the model structure is shown in Figure 4 Figure.

[0102] In S250, the intelligent prediction of the turbomachinery support stiffness using the intelligent prediction model of turbomachinery support stiffness includes:

[0103] Performing multi-scale feature extraction and transformation on the newly input support stiffness and damping ratio data to obtain a transformation result;

[0104] Performing a transformation operation on the transformation result to obtain a multi-scale feature vector;

[0105] Performing feature fusion on the multi-scale feature vector to obtain the intelligent prediction result of the turbomachinery support stiffness;

[0106] Among them, the intelligent prediction result of the turbomachinery support stiffness includes the dynamic stiffness curves of the support structure in the vertical and horizontal directions.

[0107] Multi-scale feature extraction and transformation can effectively capture the global characteristics and local details in the input data, ensuring that features at different scales are fully characterized. It can more comprehensively reflect the variation law of the support stiffness and damping ratio, providing rich basic information for subsequent feature processing. In addition, multi-scale feature extraction can enhance the generalization ability of the model, enabling it to adapt to different working conditions.

[0108] Converting the transformation result into a multi-scale feature vector can realize the structured representation of complex data. This process effectively maps the original features to a high-dimensional feature space, helps to reveal the deep-seated associations between data, enhances the model's ability to capture non-linear relationships, and thus provides a more accurate data basis for intelligent prediction.

[0109] Feature fusion can integrate the key information in multi-scale features, eliminate redundancy, enhance the integrity and stability of feature expression, contribute to improving the accuracy and robustness of the prediction model, enable the model to make full use of the synergistic effect of different-scale features, and thus generate more reliable prediction results.

[0110] By generating the dynamic stiffness curves of the support structure in the vertical and horizontal directions, the dynamic characteristics of the turbomachinery support stiffness can be comprehensively presented. These curves provide an important basis for the monitoring and analysis of the equipment operating state, help to detect abnormalities in a timely manner, and improve the safety and stability of equipment operation. In addition, these results can also be used for optimizing design, formulating maintenance strategies, and improving operating performance.

[0111] In some embodiments, performing multi-scale feature extraction and transformation on the newly input support stiffness and damping ratio data to obtain a transformation result includes:

[0112] Extracting different-scale features of the newly input support stiffness and damping ratio data;

[0113] Using the non-linear variations of multiple activation functions, convert the feature extraction results of each scale of the newly input support stiffness and damping ratio data into a new representation;

[0114] Perform weighted summation and non-linear transformation on the new representation to obtain a transformation result.

[0115] Here, using the non-linear variations of multiple activation functions to convert the feature extraction results of each scale of the newly input support stiffness and damping ratio data into a new representation can be understood as:

[0116] For a set of input data , at each scale during the feature extraction process, the first layer neurons of MKAN convert each input through the non-linear variations of multiple activation functions into a new representation. Specifically, for the th input , through the processing of the th neuron, we can obtain , that is:

[0117] ;

[0118] wherein, represents the newly formed representation, is the activation function and , represents the weight of the th neuron in the n th layer connecting the newly input support stiffness and damping ratio data l , is the corresponding bias, represents the value of the th B-spline basis function at , N is the number of B-spline basis functions, are the control points of the spline basis function, represents the grid size at scale s , represents the spline order of the B-spline curve at scale s .

[0119] Perform a transformation operation on the transformation result to obtain a multi-scale feature vector including:

[0120] By forming a univariate continuous function for the input data, and transforming to a new feature space through non-linear mapping. Further, the second layer grid performs weighted summation and non-linear transformation on the output of the first layer to obtain :

[0121] ;

[0122] After multi-scale feature extraction, the input data is transformed into a multi-scale feature vector through a concatenation operation, and its expression is as follows:

[0123] ;

[0124] Wherein, represents the conversion result obtained at the scale and , is an activation function and , represents connecting the support stiffness and damping ratio data of the new input of the n layer, the weight of the second neuron, is the corresponding bias, represents the th B-spline basis function at value, N is the number of B-spline basis functions, is the control point of the spline basis function, represents the scale s under the grid size, represents the scale s under the spline order of the B-spline curve.

[0125] Finally, through non-linear weighted summation of the multi-scale features feature fusion can be performed to obtain the final output :

[0126] ;

[0127] Wherein, represents the intelligent prediction result of the turbomachinery support stiffness, represents the weight of the fusion layer, represents the multi-scale feature vector, represents the bias of the fusion layer.

[0128] Combining the above formula and the Kolmogorov-Arnold theorem, the approximate principle of MKAN can be obtained as follows:

[0129] ;

[0130] On this basis, the present invention designs the loss function of the intelligent prediction model for the turbomachinery support stiffness including:

[0131] ;

[0132] Among them, represents the number of samples, represents the loss function used to measure the difference between the predicted value and the true value, represents the network model, represents the scale s at which the features of the m th sample are input, represents the corresponding true value.

[0133] Based on the above technical solutions, the present invention also verifies the intelligent prediction model for the support stiffness of turbomachinery, including:

[0134] Comparing different models, including Convolutional Neural Network (CNN), Residual Network (ResNet), and a new neural network architecture (Kolmogorov - Arnold Networks, KAN). The KAN model was tested at three different scales, which correspond to the scales used in the MKAN model, namely grid sizes of 5, 10, and 20. The results were verified by the Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), and Coefficient of Determination (R 2 ). All three models used the Adam optimizer for parameter optimization. The training process was iterated 500 times, with a learning rate of 0.001 and a batch size of 16.

[0135] The formulas for MAE, RMSE, and R 2 Coefficient of Determination are:

[0136] ; ; ;

[0137] Among them, represents the number of samples in one batch of training, represents the true values of the samples in one batch, represents the predicted values of the samples in one batch, represents the th sample, represents the total number of samples.

[0138] The loss during the training process reflects the convergence ability of the model. Figure 5 The loss changes of the experimental model under vertical and horizontal training data are as follows. As Figure 5 (a) and Figure 5 (b) show, the convergence and stability of the MKAN model are significantly better than those of other models, with lower loss values. Specifically, the loss curve of the MKAN model shows a rapid downward trend and tends to be stable in fewer iterations, finally reaching the lowest loss value. This indicates that MKAN has enhanced feature learning and optimization capabilities during the training process. In contrast, the ResNet model shows a higher initial loss and a slower descent rate. Although it finally tends to be stable, its final loss is still significantly greater than that of the MKAN model, indicating its weaker ability to capture data features. The CNN model performs slightly better than ResNet; however, its final loss is still significantly higher than that of MKAN, indicating its relatively limited generalization performance. Although the KAN model shows a faster convergence rate in the vertical direction, it does not exceed MKAN, which strengthens the assertion of enhanced feature capture through multi-scale extraction. In addition, although the KAN model converges rapidly in the vertical direction, its training process lacks smoothness, especially in the horizontal direction, where there is a large peak, indicating a certain degree of instability.

[0139] For the prediction of dynamic stiffness in the vertical direction, the present invention first evaluates the overall generalization of the model. Figure 6 is the evaluation factor for each model in any vertical direction verification experiment. As Figure 6 (a) shows, MKAN is superior to other models, with higher consistency and stability, and maintains a high R 2 score under all test conditions. As can be seen from Figure 6 (b), MKAN maintains the lowest MAE value, reflecting its ability to minimize prediction errors. KAN also achieves low error values in most cases, but lags slightly behind MKAN in some areas. In contrast, the MAE values of ResNet and CNN are higher, and the errors of some data points are significantly larger. As can be seen from Figure 6 (c), MKAN can obtain the lowest RMSE value in all test data, further confirming its superior error control ability. Although KAN is superior to ResNet and CNN, it still lags behind MKAN.

[0140] The quantitative evaluation and summary of the prediction results are shown in Table 1. Table 1 presents the average evaluation factors of the prediction results of each model in the verification experiments in any vertical direction. Obviously, the MKAN model outperforms other models in all evaluation metrics. Specifically, its average coefficient of determination (Average Coefficient of Determination, Mean R2 score) is 0.963401, which is 9.17%, 3.98%, 4%, 2.02%, and 3.74% higher than that of ResNet, CNN, and KAN under different grid sizes respectively. In terms of the mean absolute error (Mean Absolute Error, Mean MAE), the error of the MKAN model is the lowest, at 0.025573, which is 49.44% lower than that of the ResNet model, 39.66% lower than that of the CNN model, and 30.78%, 16.58%, and 32.24% lower than that of the KAN model with different grid sizes. This indicates that MKAN significantly reduces the prediction error and more accurately estimates the true value in the real prediction scenario. In terms of the mean root mean square error (Mean Root Mean Square Error, MeanRMSE), the error of MKAN is 0.037957, which is 48.92%, 37.51%, 31.78%, 21.33%, and 29.59% lower than that of the ResNet, CNN, and KAN models with different grid sizes respectively. The significant reduction in RMSE indicates that MKAN provides better error control, especially when dealing with large deviations, and maintains higher prediction stability over a wide range of data.

[0141] Table 1 Schematic table of average evaluation factors for verification experiments in each vertical direction

[0142]

[0143] In addition, the present invention uses specific sample predictions to accurately evaluate the effectiveness and superiority of the network model. Figure 7 are the prediction results of each model in Experiment 1. In Figure 7 (a), there are obvious deviations in the prediction results of CNN, especially in the medium and high frequency bands, and it cannot accurately capture the dynamic stiffness characteristics, especially the peak value. The prediction results of the ResNet model are as Figure 7 (b) shows that it provides improved performance, especially in the range of 0.6 to 0.8, but there are still errors, especially when capturing the peak value. Figure 7(c)-(e) show the performance of the KAN model under different grid sizes. They perform well in the low-frequency band and some mid-frequency bands; however, they deviate from the true values in the high-frequency region. It has been observed that using a finer grid does not enhance the generalization of the model; on the contrary, it only improves the accuracy in a specific fitting region. In addition, this is accompanied by an increase in local noise. In contrast, as Figure 7 shown in (f), the MKAN model exhibits good fitting ability across the entire frequency band. It is worth noting that MKAN can accurately capture the dynamic stiffness curve and identify multiple frequency peaks.

[0144] The dynamic stiffness data in the horizontal direction is evaluated Figure 8 for the comparison of the prediction performance of each model. It is obvious that the performance of MKAN is significantly better than that of other models. In terms of the R 2 coefficient, MKAN always provides the best performance in all sample experiments, while achieving the lowest levels of MAE and RMSE in almost all cases. This highlights MKAN's excellent ability to comprehensively capture complex patterns in the input data.

[0145] The present invention calculates the statistical average of the evaluation metrics for all sample data sets, as shown in Table 2. From these results, it can be seen that the MKAN model outperforms other models in all metrics. Its average R 2 score reaches 0.994681, which is about 7.17% higher than that of ResNet, 2.99% higher than that of CNN, and 1.57%, 1.09%, 1.03% higher than that of KAN. In terms of the average MAE, MKAN has the smallest error of 0.008656, which is 81.46% lower than that of ResNet, 74.51% lower than that of CNN, and 60.15%, 57.24% and 57.10% lower than that of the three-parameter KAN, respectively. Similarly, for the average RMSE, MKAN performs excellently with an error of 0.015608, which is 77.10% lower than that of ResNet, 67.75% lower than that of CNN, and 52.06%, 46.00% and 46.45% lower than that of the three-parameter KAN, respectively. In summary, these results once again verify the superiority and effectiveness of the MKAN model for dynamic stiffness prediction.

[0146] Table 2 Schematic table of average evaluation factors for each horizontal direction verification experiment

[0147]

[0148] In addition, the present invention also examines specific experimental cases to evaluate the effectiveness of the network model prediction. Figure 9The prediction results of each model in Experiment 2. The CNN model effectively captures the overall trend of the dynamic stiffness in most frequency ranges. However, it shows significant deviations, especially in the higher frequency range from 0.7 to 1.0. The ResNet model shows superior performance in the frequency range of 0.4 - 0.8. However, in the higher frequency range, especially near 0.9, it still shows inaccuracies. For the three mesh sizes, the KAN model shows effective performance in the frequency range of 0.3 - 0.7. However, it shows deviations in all high-frequency bands, especially from 0.9 to 1.0. In contrast, the MKAN model performs better than other models and is highly consistent with the true values in almost all frequency intervals. It accurately captures the fluctuation characteristics, especially in the higher frequency regions. Overall, the MKAN model shows excellent performance throughout the frequency range, with excellent prediction accuracy and adaptability to dynamic characteristics, while other models show limitations in the high-frequency range.

[0149] According to the second aspect of the embodiments of the present invention, there is also provided an intelligent prediction device for the support stiffness of a turbomachine, referring to Figure 10 as shown, the intelligent prediction device 1000 for the support stiffness of a turbomachine includes:

[0150] A model establishment module 1010, configured to establish a reduced finite element model based on a true turbomachine foundation support model;

[0151] A sample set construction module 1020, configured to construct a data sample set of the support stiffness and damping ratio of the reduced finite element model by using the Latin hypercube sampling method;

[0152] A curve generation module 1030, configured to perform finite element analysis on each data sample in the data sample set to generate dynamic stiffness curves of the support structure in the vertical and horizontal directions;

[0153] A prediction model generation module 1040, configured to construct an intelligent prediction model for the support stiffness of a turbomachine according to the data sample set and the dynamic stiffness curves of the support structure in the vertical and horizontal directions;

[0154] An implementation prediction module 1050, configured to perform intelligent prediction of the support stiffness of a turbomachine by using the intelligent prediction model for the support stiffness of a turbomachine.

[0155] It should be noted that although several modules and sub - modules of the intelligent prediction device for the support stiffness of turbomachinery are mentioned in the above - detailed description, this division is not mandatory. In fact, according to the embodiments of the present invention, the features and functions of two or more of the above - described modules or sub - modules can be embodied in one module or unit. Conversely, the features and functions of one module or sub - module described above can be further divided and embodied by multiple modules or sub - modules.

[0156] In addition, in an exemplary embodiment of the present invention, an electronic device capable of implementing the method of the above - mentioned intelligent prediction method for the support stiffness of turbomachinery is also provided.

[0157] Those skilled in the art can understand that various aspects of the present invention can be implemented as a system, a method, or a program product. Therefore, various aspects of the present invention can be specifically implemented in the following forms: a complete hardware embodiment, a complete software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, which can be collectively referred to as "circuit", "module", or "system" here.

[0158] The following refers to Figure 11 to describe the electronic device 1100 according to this embodiment of the present invention. Figure 11 The illustrated electronic device 1100 is only an example and should not impose any limitation on the functions and usage scope of the embodiments of the present invention.

[0159] As Figure 11 shown, the electronic device 1100 is presented in the form of a general - purpose computing device. The components of the electronic device 1100 may include, but are not limited to: at least one of the above - mentioned processing units 1110, at least one of the above - mentioned storage units 1120, a bus 1130 connecting different system components (including the storage unit 1120 and the processing unit 1110), and a display unit 1140.

[0160] Among them, the storage unit stores program code, and the program code can be executed by the processing unit 1110, so that the processing unit 1110 executes the steps according to various exemplary embodiments of the present invention described in the above - mentioned "exemplary method" part of the present invention. For example, the processing unit 410 can execute as Figure 2S210, as shown, based on a true turbomachinery foundation support model, establish a reduced finite element model; S220, using the Latin hypercube sampling method, construct a data sample set of the support stiffness and damping ratio of the reduced finite element model; S230, perform finite element analysis on each data sample in the data sample set to generate dynamic stiffness curves of the support structure in the vertical and horizontal directions; S240, according to the data sample set and the dynamic stiffness curves of the support structure in the vertical and horizontal directions, construct an intelligent prediction model for the turbomachinery support stiffness; S250, use the intelligent prediction model for the turbomachinery support stiffness to perform intelligent prediction of the turbomachinery support stiffness.

[0161] The storage unit 1120 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 1121 and / or a cache storage unit 1122, and may further include a read-only storage unit (ROM) 1123.

[0162] The storage unit 1120 may also include a program / utilities 1124 having a set (at least one) of program modules 1125. Such program modules 1125 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment.

[0163] The bus 1130 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus structures.

[0164] The electronic device 1100 may also communicate with one or more external devices 1170 (such as a keyboard, a pointing device, a Bluetooth device, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 1100, and / or may communicate with any device that enables the electronic device 1100 to communicate with one or more other computing devices (such as a router, a modem, etc.). Such communication may be through an input / output (I / O) interface 1150. And, the electronic device 1100 may also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 1160. As shown, the network adapter 1160 communicates with other modules of the electronic device 1100 through the bus 1130. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 1100, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0165] From the description of the above embodiments, those skilled in the art can easily understand that the exemplary embodiments described herein can be implemented by software or by a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, a USB flash drive, a portable hard drive, etc.) or on a network, including several instructions to enable a computing device (such as a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present invention.

[0166] In an exemplary embodiment of the present invention, there is also provided a computer-readable storage medium having stored thereon a program product capable of implementing the above method of the present invention. In some possible embodiments, various aspects of the present invention can also be implemented in the form of a program product, which includes program code. When the program product runs on a terminal device, the program code is used to cause the terminal device to execute the steps according to various exemplary embodiments of the present invention described in the above "Exemplary Method" section of the present invention.

[0167] Referring Figure 12 As shown, a program product 1200 for implementing the above intelligent prediction method for the foundation stiffness of a turbomachine according to an embodiment of the present invention is described. It can be a portable compact disc read-only memory (CD-ROM) and includes program code, and can run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto. In the present invention, the readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or in combination with an instruction execution system, apparatus, or device.

[0168] The program product can adopt any combination of one or more readable storage media. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0169] The program code for performing the operations of the present invention can be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, executed as a stand-alone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or, alternatively, can be connected to an external computing device (e.g., by using an Internet service provider to connect through the Internet).

[0170] In addition, the above-mentioned drawings are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present invention, rather than for limiting purposes. It is easy to understand that the processes shown in the above-mentioned drawings do not indicate or limit the chronological order of these processes. Additionally, it is also easy to understand that these processes can be executed, for example, synchronously or asynchronously in multiple modules.

[0171] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software, or can be implemented by a combination of software and necessary hardware. Therefore, the technical solution according to the embodiments of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiments of the present invention.

[0172] After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily conceive of other embodiments of the present invention. The present invention is intended to cover any variations, uses, or adaptations of the present invention, which follow the general principles of the present invention and include well-known common knowledge or conventional technical means in the technical field not disclosed by the present invention. The specification and embodiments are only regarded as exemplary, and the true scope and spirit of the present invention are pointed out by the claims.

[0173] It should be understood that the present invention is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. A method for intelligent prediction of turbomachinery support stiffness, characterized in that: include: Based on the real turbomachinery foundation support model, a reduced finite element model is established; Using a Latin hypercube sampling method, a data sample set of support stiffness and damping ratio of the reduced finite element model is constructed; Performing finite element analysis on each data sample in the data sample set to generate dynamic stiffness curves of the support structure in vertical and horizontal directions; Constructing a turbomachinery support stiffness intelligent prediction model based on the data sample set and the dynamic stiffness curves of the support structure in vertical and horizontal directions; Using the turbine machinery support stiffness intelligent prediction model to perform turbine machinery support stiffness intelligent prediction; Using the turbine machinery support stiffness intelligent prediction model to perform turbine machinery support stiffness intelligent prediction includes: Perform multi-scale feature extraction and conversion on the newly input support stiffness and damping ratio data to obtain the conversion result; Performing a transformation operation on the transformation result to obtain a multi-scale feature vector; Performing feature fusion on the multi-scale feature vectors to obtain an intelligent prediction result of the turbomachinery support stiffness; Wherein, the intelligent prediction result of the turbomachinery support stiffness includes the dynamic stiffness curves of the support structure in the vertical and horizontal directions.

2. The method for intelligent prediction of turbomachinery support stiffness according to claim 1, characterized in that: The newly input support stiffness and damping ratio data are subjected to multi-scale feature extraction and conversion, and the conversion results include: Extracting different scale features of the newly input support stiffness and damping ratio data; The feature extraction results of each scale of the newly input support stiffness and damping ratio data are converted into a new representation by utilizing nonlinear changes of multiple activation functions; The new representation is subjected to weighted summation and nonlinear transformation to obtain a transformation result.

3. The intelligent prediction method for turbomachinery support stiffness according to claim 2, characterized in that: Using nonlinear changes of multiple activation functions, the feature extraction results of each scale of the newly input support stiffness and damping ratio data are converted into a new representation including: in, The new representation formed by the representation transformation, is the activation function and , Indicates the newly entered support stiffness and damping ratio data for the connection No. n Layer l The weight of the neuron, is the corresponding bias, Indicates The B-spline basis functions are The value of N is the number of B-spline basis functions, are the control points of the spline basis function, Representation scale The grid size below, Representation scale The spline degree of the B-spline curve below.

4. The method for intelligent prediction of turbomachinery support stiffness according to claim 1, characterized in that: The transformation result is transformed to obtain a multi-scale feature vector include: in, Indicated in The conversion result obtained by scale is , is the activation function and , Indicates the newly entered support stiffness and damping ratio data for the connection No. n The weight of the second neuron in the layer, is the corresponding bias, Indicates The B-spline basis functions are The value of N is the number of B-spline basis functions, are the control points of the spline basis function, Representation scale The grid size below, Representation scale The spline degree of the B-spline curve below.

5. The intelligent prediction method for turbomachinery support stiffness according to claim 1, characterized in that: The multi-scale feature vectors are subjected to feature fusion to obtain the intelligent prediction results of the turbomachinery support stiffness, including: in, It represents the intelligent prediction result of turbomachinery support stiffness. represents the weight of the fusion layer, represents the multi-scale feature vector, Represents the bias of the fusion layer.

6. The method for intelligent prediction of turbomachinery support stiffness according to any one of claims 1 to 5, characterized in that: The loss function of the intelligent prediction model for turbomachinery support stiffness include: in, represents the number of samples, represents the loss function used to measure the difference between the predicted value and the true value, Represents the network model, Representation scale s Next input m The characteristics of the samples, Indicates the corresponding true value.

7. A device based on the turbomachinery support stiffness intelligent prediction method according to any one of claims 1 to 6, characterized in that: include: Model building module for building reduced finite element models based on real turbomachinery foundation support models; A sample set construction module, used to construct a data sample set of support stiffness and damping ratio of the reduced finite element model by using a Latin hypercube sampling method; A curve generating module, used for performing finite element analysis on each data sample in the data sample set to generate dynamic stiffness curves of the supporting structure in vertical and horizontal directions; A prediction model generation module, used for constructing a turbomachinery support stiffness intelligent prediction model based on the data sample set and the dynamic stiffness curves of the support structure in the vertical and horizontal directions; A prediction module is implemented to perform intelligent prediction of turbine machinery support stiffness using the turbine machinery support stiffness intelligent prediction model.

8. An electronic device, characterized in that: include: processor; as well as A memory having computer-readable instructions stored thereon, wherein the computer-readable instructions, when executed by the processor, implement the method for intelligent prediction of support stiffness of a turbomachinery according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the method for intelligently predicting the support stiffness of a turbomachinery according to any one of claims 1 to 6 is implemented.

Images